How to stress-test a DeFi wallet without making a trade
Replay past market shocks on your supported holdings to see how they would have re-priced — a read-only, historical stress test, not a forecast and not a trade. Non-custodial, not investment advice.
A stress test answers a narrow, useful question: if a past market shock happened again to the assets you hold today, how would those positions have re-priced? Routescore runs that replay against your supported holdings without you signing, sending, or moving anything — it is read-only and non-custodial from the first click to the last. What comes back is a modeled, historical number, not a forecast and not advice.
What a historical stress test is
A historical stress test takes a real event from the past — a flash crash, a stablecoin depeg, an exchange collapse — and replays that event's price path against the supported holdings in your wallet right now. It is a backward-looking "what would this have done to me" exercise, not a forward-looking "what will happen next" claim.
The scenarios are fixed, named historical windows. At launch they include:
- ETH flash crash — the March 2020 COVID-19 liquidation cascade.
- LUNA / UST collapse — the May 2022 algorithmic-stablecoin death spiral.
- FTX contagion — the November 2022 exchange-insolvency forced selling.
- USDC depeg — the March 2023 SVB-exposure depeg to roughly $0.88.
Each scenario re-prices your current supported positions through the drawdown that actually occurred in that window. The output is modeled and point-in-time: it reflects the historical move and your present balances, not a probability that the move repeats.
This feature replays history. It does not predict the next shock, and it does not forecast returns. A scenario showing a large modeled drawdown is not a warning that it will happen; a scenario showing a small one is not a promise of safety.
How to run it read-only without making a trade
Stress scenarios live on the Pro tier and reuse the same supported-wallet pricing the rest of the workbench uses, so there is nothing new to connect.
- Point the workbench at a wallet. Connect a wallet or use one you have already saved. This is read-only: connecting never grants a spend approval and never needs a private key or seed phrase.
- Pick a historical scenario. Choose one of the named windows above from the scenario picker.
- Run the scenario. The workbench re-prices your supported holdings through that window and returns a modeled impact.
Running a scenario never creates a transaction. Routescore is read-only and non-custodial: it does not execute trades, route funds, or hold your assets, and it never asks for a key or seed phrase. If your wallet has no supported priced rows yet, the workbench falls back to a clearly labeled demo portfolio so you can still see exactly how the mechanic works before connecting your own.
What "supported holdings" covers and what is excluded
A stress test is only as honest as its coverage, so Routescore is explicit about what it can and cannot price. Supported holdings are the high-volume base assets and major ETH staking/restaking exposures that have reliable pricing for the historical windows.
Supported (priced and stress-tested):
- Native ETH balances.
- WBTC.
- Major stablecoins (USDC, USDT, DAI).
- Major ETH staking / restaking exposures (e.g. stETH, wstETH).
Excluded (not priced into the scenario):
- Long-tail and low-liquidity tokens not yet in the supported set.
- Unpriced tokens — they may show in your wallet but carry no scenario value.
- NFTs and other illiquid positions.
- Cross-chain bridge exposure, which is surfaced separately and not modeled in the scenario.
- Anything without a reliable price for the historical window being replayed.
If a position is excluded, it simply does not contribute to the modeled impact — the number you read covers your supported value only, and the workbench labels how much of your wallet that represents.
How to read the modeled change + per-position breakdown
The result is built to be read top-down, from the headline number to the position that drove it:
- Estimated impact — the modeled change to your supported value, shown in USD (for example, −$1,240). It is the re-priced value through the historical window, not a realized loss and not a prediction.
- Share of supported value — the same impact expressed as a percent of your supported holdings, so a dollar figure stays in context.
- Scenario risk score — a 0–1 tag for how severe the historical window was, to compare one scenario against another.
- Modeled impact by position — a chart of how the value moved across the window, with the shock event marked.
- Affected positions — a per-position breakdown listing each supported holding, its modeled USD impact, and its modeled drawdown percent, so you can see whether one protocol or pair is carrying most of the move.
Every figure here is modeled and point-in-time. The breakdown tells you where the modeled impact concentrated in a past event — it does not tell you what your wallet will do next.
Not advice, and not a prediction
Routescore is decision support, not investment, legal, or tax advice. It does not tell you what to buy or sell, and it does not name a position to exit. A low modeled impact is not a guarantee of safety, and a high one is not a forecast that the shock will recur — both are simply replays of history against your current supported holdings.
Stress-scenario outputs are modeled, point-in-time decision support and not investment advice. Nothing in DeFi is free of risk, and a historical replay cannot promise a future outcome. Use the number to reason about your own exposure, not as a recommendation.
Keep a record
The most durable value comes from writing down what you saw and why it mattered. After a run you can:
- Save the scenario to your hash-recorded journal as a tamper-evident entry, capturing the assumptions and the modeled result you actually reviewed.
- Copy the scenario JSON — a portable artifact with the scenario id, your supported value, the modeled impact, and the per-position breakdown.
- Copy the scenario link to revisit or share the exact same historical window.
Recording the decision turns a one-off read into decision memory: the next time this market gets loud, you can see what you concluded last time and on what basis — all without a single trade.